Research Article
Real-Time Human Ear Detection Based on the Joint of Yolo
and RetinaFace
Huy Nguyen Quoc and Vinh Truong Hoang
Ho Chi Minh City Open University, 35-37 Ho Hao Hon Street, Ward Co Giang, District 1, Ho Chi Minh City, Vietnam
CorrespondenceshouldbeaddressedtoVinhTruongHoang;vinh.th@ou.edu.vn
Received 21 September 2021; Revised 14 October 2021; Accepted 18 October 2021; Published 8 November 2021
AcademicEditor:BaltazarAguirreHernandez
Copyright © 2021 Huy Nguyen Quoc and Vinh Truong Hoang. is is an open access article distributed under the Creative
Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.
Biometrictraitsgraduallyprovedtheirimportanceinreal-lifeapplications,especiallyinidentificationfield.Amongtheavailable
biometrictraits,theuniqueshapeofthehumanearhasalsoreceivedloadsofattentionfromscientiststhroughtheyears.Hence,
numerousear-basedapproacheshavebeenproposedwithpromisingperformance.Withthesemethods,plentyproblemscanbe
solvebythedistinctivenessofearfeatures,suchasrecognizinghumanwithmaskordiagnoseear-relateddiseases.Asacomplete
identificationsystemrequiresaneffectivedetectorforreal-timeapplication,andthecurrentrichnessandvarietyofeardetection
algorithmsarepoorduetothesmallandcomplexshapeofhumanears.Inthispaper,weintroduceanewhumaneardetection
pipelinebasedontheYOLOv3detector.Awell-knownfacedetectornamedRetinaFaceisalsoaddedinthedetectionsystemto
narrowtheregionsofinterestandenhancetheaccuracy.eproposedmethodisevaluatedonanunconstraineddataset,which
shows its effectiveness.
1. Introduction
Identificationalwaysholdsanessentialroleinourdailylives,
such as information security, banking transactions, and
e-commerce. With the development of computer vision,
most identification systems are now based on biometric
traits. However, due to the COVID-19 pandemic, people
havetowearmasksorprotectivegearsallthetimeinpublic.
isissuelimitsthepossibilityofseveralbiometricpatterns,
includingface,iris,andfingerprints.erefore,weproposed
toapplythehumaneartosubstitutetheavailablebiometric
traitsinidentificationtasks.Asahumanhearingorgan,the
earshavebeenprovedtobeasdistinctiveasotherbiometric
patterns. Specifically, parts such as the helix, the antihelix,
the tragus, the antitragus, and the fossa have formed nu-
merous curves during ear development [1]. ese curves
createtheouteroftheear,whichisalsocalledthepinna,and
providetheuniquenessofthehumanear[2].Evenearsfrom
the same person still have several differences. With these
studies, the first human ear identification system was pre-
sentedbyManuelZimberoffin1963.Afterthat,loadsofear-
basedapproacheshavebeenproposedinordertoreplacethe
common biometric traits with the human ear in several
computervisiontasksorjustsimplycombiningthefeatures
of the human ear with other biometric patterns to enhance
theperformance.Forexample,Alshazlyetal.combineddeep
learning and transfer learning models to analyze and rec-
ognize human ears [3]. Hassaballah et al. extracted features
fromearimageusingtheLBPdescriptoranditsvariantsfor
classification [4]. In 2020, Alshazly et al. proposed a neural
network to recognize unconstrained ear images [5]. In that
year, Ganapathi et al. presented a geometric feature for 3D
ear recognition [6]. Several ear comparative studies and
surveys were also made by Pflug et al. for research purpose
[7, 8]. ese approaches allow us to build multiple appli-
cationstosolveear-relatedtasks.Currently,oneofthemost
urgent and essential problems which is face with mask
recognitioncanbesolvedwitheardetectionbecauseearsare
not occluded when wearing mask. Ear recognition is also
helpful when identifying person from other angles which is
very useful for large-scale recognition tasks and cameras
with fixed angle. Furthermore, ear detection can be applied
Hindawi
Complexity
Volume 2021, Article ID 7918165, 11 pages
https://doi.org/10.1155/2021/7918165